Finds and fixes syntax errors. This is experimental software.
NOTE: Please use the saner2018 tag if you're looking to replicate the results from our SANER 2018 paper. Find the replication data on archive.org See Citation for how to cite this paper. Use the following to clone Sensibility with the correct branch:
git clone --branch=saner2018 https://github.com/naturalness/sensibility.git
Sensibility requires:
- Python 3.6.
For fixing Java files:
- Java 8 SE
For fixing JavaScript files:
- Node.JS >= 6.0 and ZeroMQ
(macOS:
brew install zeromq; Ubuntu:apt install libzmq-dev)
Researchers that wish to mine more repositories or languages will require:
- A running Redis server.
Researchers wishing to evaluate the current methods will require:
Activate a virtualenv, if that's your thing. Then,
pip install -e .
Once installed, there's one entry point to all the scripts and utilities included:
sensibility <SUBCOMMAND>
Most subcommands require the specification of a language,
with the -l <LANGUAGE> option before the subcommand.
For example, to train Java models:
sensibility -l java train-list --help
The following diagram visualizes the data flow.
Subcommands to sensibility are in black rectangles; white ovals are products.
Please contact the author to obtain the mistake database,
which is only applicable when evaluating the Java models.
To run the tests, install tox using Pip, then run tox.
You must create a GitHub OAuth token and save it as
.tokenin the repository root.Run
redis-serveron localhost on the default port.Use
sensibility mine find-reposto get a list of the top ~10k repos:sensibility mine find-repos javascript | sort -u > javascript-repos.txt
Use
bin/enqueue-repoto enqueue repos to download:sensibility mine enqueue-repo < javascript-repos.txt
Start one or more downloaders. These will dequeue a repo from the running Redis server and download sources:
sensibility mine download
Type make experiments to train all of the models and evaluate each one.
See libexec/experiments for more details.
If you use Sensibility in academic works, please use the following citation:
@inproceedings{santos2018,
author={Santos, Eddie Antonio and
Campbell, Joshua Charles and
Patel, Dhvani and
Hindle, Abram and
Amaral, Jos{\'e} Nelson},
booktitle={2018 {IEEE} 25th International Conference on Software Analysis, Evolution and Reengineering ({SANER})},
title={Syntax and {Sensibility}: Using Language Models to Detect and Correct Syntax Errors},
year={2018},
month={Mar}}
Copyright 2016, 2017 Eddie Antonio Santos [email protected]
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent